"Michael L. Hitch" wrote:

> On Tue, 7 Dec 1999, Jamie Scuglia wrote:
> 
> > On Tue, Dec 07, 1999 at 12:03:05PM +1100, Simon Burge wrote:
> > 
> > > I've just placed a -current snapshot on ftp.netbsd.org in the
> > > pub/NetBSD/arch/pmax/snapshot/1999-12-06 directory.  It should be
> > > complete except for the X11 sets - I'm building them now.
> ...
> > The updated kernel and base work fine on a DEC 5000/240 (3MAXPLUS), but
> > when we try to use it on a DEC 5000/200 (3MAX), the kernel just hangs
> > before the NetBSD copyright messages - i.e. the last message we see
> > on the screen is
> > 
> > 	[ preserving 134836 bytes of netbsd ELF symbol table]
> > 
> > Then the machine hangs - no more messages, doesn't respond to key presses.
> 
>   I was unable to boot a -current kernel on my 5000/200 running with a
> serial console.  Mine hung when it configured the serial driver.
> 
>   It looks to me like the separation between the 3100 and 5000/200 serial
> console initialization was removed, and the 5000/200 is trying to use the
> 3100 address of the serial chip.  I hacked the serial console
> initialization to use the 5000/200 address, and was able to boot the
> system (but that will break the 3100 console).  I haven't figured out what
> needs to be done to fix this yet.

Ahh, that was the KN01_SYS_DZ thing?  I'd forgotten all about it.  I
wonder if we should just case on systype in dc_ds_consinit()?  Seems
a little ugly though but this early on the address of what dc we're
using doesn't seem to be known because autoconfig hasn't happened yet
(correct?)...

Simon.